Hi Friends, I'm hoping you can help with a modeling conundrum.    I'm modeling some cost data using generalized linear models (PROC GENMOD). I have multiple observations per person, so I've included the REPEATED statement to account for the correlation between values for the same patient. However, I'm a little confused by the results. As I understand it, including a repeated statement shouldn't change the model estimates, just the CIs/SEs. But I noticed that changing the covariance structure from compund symmetric to autoregressive DOES change the estimates (and by a non-negligible amount!). This is making me wonder if I'm doing something wrong in my code... I've pasted my code below. Aany insight into why I might be getting unexpected changes in the model estimate as a result of changing the covariance structure??   proc genmod data=forreg  ; class id bg year / param=ref ref=first; model pay = bg year bg*year/ link=log dist=gamma; repeated subject=id /type=cs ; run;  ... View more

I'm hoping someone can help me understand when it is appropriate to add a WEIGHT statement to PROC MEANS versus when do I need to use PROC SURVEYMEANS. I have read the documentation and a few papers on the topic, but I still can't quite get a handle on this for my specific situation(s). I have little experience with survey data, so I apologize in advance for my ignorance.   Situation #1: I'm using an insurance claims database and I'd like to standardize my estimates to the U.S. population using weights I have calculated from U.S. census data. In other words, I will apply weights such that the demographic distribution of people in my dataset matches the demographic distribution of the U.S. and no subpopulations are over- or under-represented. There was no complex sampling scheme for this database -- it is a convenience sample. Is using a WEIGHT statement with PROC MEANS sufficient if I want to calculate the weighted average of, say, healthcare expenditures, provider visits, etc.? Why or why not?    Situation #2: Using the same database consisting of a convenience sample of the U.S. population, I want to propensity-score weight 2 treatment groups. I have calculated weights such that the 2 groups will be similar on important baseline characteristics when the weights are applied. I want to compare outcomes (eg, healthcare expenditures, provider visits, ... etc)  between the weighted groups.  Is using a WEIGHT statement with PROC MEANS sufficient, or do I need to use SURVEYMEANS.    I suspect using a WEIGHT statement might be sufficient for both of these purposes, but the only examples I can seem to find are situations where the WEIGHT statment is NOT sufficient. Again, apologies for my complete ignorance on this topic! Any helpful input or papers you can recommend are greatly appreciated! ... View more

Hi everyone,    I'm trying to use frequency matching/category matching to match a set of control patients to my selected case patients. That is, rather than individual (1:n) matching, I just want my control group to have similar distributions for a selected set of covariates. I don't need a white, 50-year-old, female case patient to have another exact white, 50-yr-old, female control patient -- I just want to be sure that if 50% of my cases are white, 20% are 50-year-olds, and 55% are female, those respective distributions are similar in the control group. Is there a straightforward way to match this way in SAS? I have done 1:n matching using PROC SURVEYSELECT, but I haven't been able to find any examples of performing frequency matching.    Any help would be greatly appreciated! Thanks!  ... View more